/src/jsonnet/core/string_utils.cpp

Source
/*
Copyright 2015 Google Inc. All rights reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

#include <iomanip>

#include "static_error.h"
#include "string_utils.h"

namespace jsonnet::internal {

UString jsonnet_string_unparse(const UString &str, bool single)
{
    UStringStream ss;
    ss << (single ? U'\'' : U'\"');
    ss << jsonnet_string_escape(str, single);
    ss << (single ? U'\'' : U'\"');
    return ss.str();
}

UString jsonnet_string_escape(const UString &str, bool single)
{
    UStringStream ss;
    for (std::size_t i = 0; i < str.length(); ++i) {
        char32_t c = str[i];
        switch (c) {
            case U'\"': ss << (single ? U"\"" : U"\\\""); break;
            case U'\'': ss << (single ? U"\\\'" : U"\'"); break;
            case U'\\': ss << U"\\\\"; break;
            case U'\b': ss << U"\\b"; break;
            case U'\f': ss << U"\\f"; break;
            case U'\n': ss << U"\\n"; break;
            case U'\r': ss << U"\\r"; break;
            case U'\t': ss << U"\\t"; break;
            case U'\0': ss << U"\\u0000"; break;
            default: {
                if (c < 0x20 || (c >= 0x7f && c <= 0x9f)) {
                    // Unprintable, use \u
                    std::stringstream ss8;
                    ss8 << "\\u" << std::hex << std::setfill('0') << std::setw(4)
                        << (unsigned long)(c);
                    ss << decode_utf8(ss8.str());
                } else {
                    // Printable, write verbatim
                    ss << c;
                }
            }
        }
    }
    return ss.str();
}

unsigned long jsonnet_string_parse_unicode(const LocationRange &loc, const char32_t *c)
{
    unsigned long codepoint = 0;
    // Expect 4 hex digits.
    for (unsigned i = 0; i < 4; ++i) {
        auto x = (unsigned char)(c[i]);
        unsigned digit;
        if (x == '\0') {
            auto msg = "Truncated unicode escape sequence in string literal.";
            throw StaticError(loc, msg);
        } else if (x >= '0' && x <= '9') {
            digit = x - '0';
        } else if (x >= 'a' && x <= 'f') {
            digit = x - 'a' + 10;
        } else if (x >= 'A' && x <= 'F') {
            digit = x - 'A' + 10;
        } else {
            std::stringstream ss;
            ss << "Malformed unicode escape character, "
               << "should be hex: '" << x << "'";
            throw StaticError(loc, ss.str());
        }
        codepoint *= 16;
        codepoint += digit;
    }
    return codepoint;
}

bool is_bmp_codepoint(const unsigned long codepoint)
{
    return codepoint < 0xd800 || (codepoint >= 0xe000 && codepoint < 0x10000);
}

char32_t decode_utf16_surrogates(const LocationRange &loc, const unsigned long high, const unsigned long low)
{
    if (high >= 0xd800 && high < 0xdc00 && low >= 0xdc00 && low < 0xe000) {
        return 0x10000 + ((high & 0x03ff) << 10) + (low & 0x03ff);
    } else {
        std::stringstream ss;
        ss << "Invalid UTF-16 bytes";
        throw StaticError(loc, ss.str());
    }
}

UString jsonnet_string_unescape(const LocationRange &loc, const UString &s)
{
    UString r;
    const char32_t *s_ptr = s.c_str();
    for (const char32_t *c = s_ptr; *c != U'\0'; ++c) {
        switch (*c) {
            case '\\':
                switch (*(++c)) {
                    case '"':
                    case '\'': r += *c; break;

                    case '\\': r += *c; break;

                    case '/': r += *c; break;

                    case 'b': r += '\b'; break;

                    case 'f': r += '\f'; break;

                    case 'n': r += '\n'; break;

                    case 'r': r += '\r'; break;

                    case 't': r += '\t'; break;

                    case 'u': {
                        ++c;  // Consume the 'u'.
                        unsigned long codepoint = jsonnet_string_parse_unicode(loc, c);

                        // Leave us on the last char, ready for the ++c at
                        // the outer for loop.
                        c += 3;
                        if (!is_bmp_codepoint(codepoint)) {
                           if (*(++c) != '\\') {
                                std::stringstream ss;
                                ss << "Invalid non-BMP Unicode escape in string literal";
                                throw StaticError(loc, ss.str());
                           }
                           if (*(++c) != 'u') {
                                std::stringstream ss;
                                ss << "Invalid non-BMP Unicode escape in string literal";
                                throw StaticError(loc, ss.str());
                           }
                           ++c;
                           unsigned long codepoint2 = jsonnet_string_parse_unicode(loc, c);
                           c += 3;
                           codepoint = decode_utf16_surrogates(loc, codepoint, codepoint2);
                       }
                       r += codepoint;
                    } break;

                    case '\0': {
                        auto msg = "Truncated escape sequence in string literal.";
                        throw StaticError(loc, msg);
                    }

                    default: {
                        std::stringstream ss;
                        std::string utf8;
                        encode_utf8(*c, utf8);
                        ss << "Unknown escape sequence in string literal: '" << utf8 << "'";
                        throw StaticError(loc, ss.str());
                    }
                }
                break;

            default:
                // Just a regular letter.
                r += *c;
        }
    }
    return r;
}

}  // namespace jsonnet::internal

Line	Count	Source
1		/*
2		Copyright 2015 Google Inc. All rights reserved.
3
4		Licensed under the Apache License, Version 2.0 (the "License");
5		you may not use this file except in compliance with the License.
6		You may obtain a copy of the License at
7
8		http://www.apache.org/licenses/LICENSE-2.0
9
10		Unless required by applicable law or agreed to in writing, software
11		distributed under the License is distributed on an "AS IS" BASIS,
12		WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13		See the License for the specific language governing permissions and
14		limitations under the License.
15		*/
16
17		#include <iomanip>
18
19		#include "static_error.h"
20		#include "string_utils.h"
21
22		namespace jsonnet::internal {
23
24		UString jsonnet_string_unparse(const UString &str, bool single)
25	1.95M	{
26	1.95M	UStringStream ss;
27	1.95M	ss << (single ? U'\'' : U'\"');
28	1.95M	ss << jsonnet_string_escape(str, single);
29	1.95M	ss << (single ? U'\'' : U'\"');
30	1.95M	return ss.str();
31	1.95M	}
32
33		UString jsonnet_string_escape(const UString &str, bool single)
34	1.95M	{
35	1.95M	UStringStream ss;
36	3.00G	for (std::size_t i = 0; i < str.length(); ++i) {
37	3.00G	char32_t c = str[i];
38	3.00G	switch (c) {
39	1.70M	case U'\"': ss << (single ? U"\"" : U"\\\""); break;
40	1.76M	case U'\'': ss << (single ? U"\\\'" : U"\'"); break;
41	2.55G	case U'\\': ss << U"\\\\"; break;
42	1.98M	case U'\b': ss << U"\\b"; break;
43	3.84M	case U'\f': ss << U"\\f"; break;
44	39.4M	case U'\n': ss << U"\\n"; break;
45	2.60M	case U'\r': ss << U"\\r"; break;
46	6.26M	case U'\t': ss << U"\\t"; break;
47	464k	case U'\0': ss << U"\\u0000"; break;
48	396M	default: {
49	396M	if (c < 0x20 \|\| (c >= 0x7f && c <= 0x9f)) {
50		// Unprintable, use \u
51	51.1M	std::stringstream ss8;
52	51.1M	ss8 << "\\u" << std::hex << std::setfill('0') << std::setw(4)
53	51.1M	<< (unsigned long)(c);
54	51.1M	ss << decode_utf8(ss8.str());
55	344M	} else {
56		// Printable, write verbatim
57	344M	ss << c;
58	344M	}
59	396M	}
60	3.00G	}
61	3.00G	}
62	1.95M	return ss.str();
63	1.95M	}
64
65		unsigned long jsonnet_string_parse_unicode(const LocationRange &loc, const char32_t *c)
66	83.5k	{
67	83.5k	unsigned long codepoint = 0;
68		// Expect 4 hex digits.
69	416k	for (unsigned i = 0; i < 4; ++i) {
70	333k	auto x = (unsigned char)(c[i]);
71	333k	unsigned digit;
72	333k	if (x == '\0') {
73	70	auto msg = "Truncated unicode escape sequence in string literal.";
74	70	throw StaticError(loc, msg);
75	333k	} else if (x >= '0' && x <= '9') {
76	267k	digit = x - '0';
77	267k	} else if (x >= 'a' && x <= 'f') {
78	22.3k	digit = x - 'a' + 10;
79	43.5k	} else if (x >= 'A' && x <= 'F') {
80	43.2k	digit = x - 'A' + 10;
81	43.2k	} else {
82	231	std::stringstream ss;
83	231	ss << "Malformed unicode escape character, "
84	231	<< "should be hex: '" << x << "'";
85	231	throw StaticError(loc, ss.str());
86	231	}
87	333k	codepoint *= 16;
88	333k	codepoint += digit;
89	333k	}
90	83.2k	return codepoint;
91	83.5k	}
92
93		bool is_bmp_codepoint(const unsigned long codepoint)
94	82.1k	{
95	82.1k	return codepoint < 0xd800 \|\| (codepoint >= 0xe000 && codepoint < 0x10000);
96	82.1k	}
97
98		char32_t decode_utf16_surrogates(const LocationRange &loc, const unsigned long high, const unsigned long low)
99	1.02k	{
100	1.02k	if (high >= 0xd800 && high < 0xdc00 && low >= 0xdc00 && low < 0xe000) {
101	980	return 0x10000 + ((high & 0x03ff) << 10) + (low & 0x03ff);
102	980	} else {
103	43	std::stringstream ss;
104	43	ss << "Invalid UTF-16 bytes";
105	43	throw StaticError(loc, ss.str());
106	43	}
107	1.02k	}
108
109		UString jsonnet_string_unescape(const LocationRange &loc, const UString &s)
110	11.8M	{
111	11.8M	UString r;
112	11.8M	const char32_t *s_ptr = s.c_str();
113	214M	for (const char32_t c = s_ptr; c != U'\0'; ++c) {
114	202M	switch (*c) {
115	1.43M	case '\\':
116	1.43M	switch (*(++c)) {
117	55.2k	case '"':
118	83.1k	case '\'': r += *c; break;
119
120	293k	case '\\': r += *c; break;
121
122	7.70k	case '/': r += *c; break;
123
124	35.3k	case 'b': r += '\b'; break;
125
126	95.9k	case 'f': r += '\f'; break;
127
128	699k	case 'n': r += '\n'; break;
129
130	50.9k	case 'r': r += '\r'; break;
131
132	88.7k	case 't': r += '\t'; break;
133
134	82.4k	case 'u': {
135	82.4k	++c; // Consume the 'u'.
136	82.4k	unsigned long codepoint = jsonnet_string_parse_unicode(loc, c);
137
138		// Leave us on the last char, ready for the ++c at
139		// the outer for loop.
140	82.4k	c += 3;
141	82.4k	if (!is_bmp_codepoint(codepoint)) {
142	1.11k	if (*(++c) != '\\') {
143	52	std::stringstream ss;
144	52	ss << "Invalid non-BMP Unicode escape in string literal";
145	52	throw StaticError(loc, ss.str());
146	52	}
147	1.06k	if (*(++c) != 'u') {
148	22	std::stringstream ss;
149	22	ss << "Invalid non-BMP Unicode escape in string literal";
150	22	throw StaticError(loc, ss.str());
151	22	}
152	1.04k	++c;
153	1.04k	unsigned long codepoint2 = jsonnet_string_parse_unicode(loc, c);
154	1.04k	c += 3;
155	1.04k	codepoint = decode_utf16_surrogates(loc, codepoint, codepoint2);
156	1.04k	}
157	82.3k	r += codepoint;
158	82.3k	} break;
159
160	8	case '\0': {
161	8	auto msg = "Truncated escape sequence in string literal.";
162	8	throw StaticError(loc, msg);
163	82.4k	}
164
165	427	default: {
166	427	std::stringstream ss;
167	427	std::string utf8;
168	427	encode_utf8(*c, utf8);
169	427	ss << "Unknown escape sequence in string literal: '" << utf8 << "'";
170	427	throw StaticError(loc, ss.str());
171	82.4k	}
172	1.43M	}
173	1.43M	break;
174
175	200M	default:
176		// Just a regular letter.
177	200M	r += *c;
178	202M	}
179	202M	}
180	11.8M	return r;
181	11.8M	}
182
183		} // namespace jsonnet::internal

Coverage Report

Created: 2025-11-24 06:35